Thanks for the thorough review of the document, Andrew, replies below.

On Tue, Jul 18, 2017 at 4:46 AM, Andrew Ayer <[email protected]> wrote:

> On page 27, the following doesn't make sense to me:
>
> > If the returned "sct" is intended to be provided to clients, then
> > "sth" and "inclusion" (if returned) SHOULD also be provided to
> > clients (e.g., if "type" was 1 then all three "TransItem"s could be
> > embedded in the certificate).
>
> First, I assume "client" means TLS client?
>
> Second, since a type of 1 means a certificate (not a precertificate),
> it's unclear how the returned TransItems could be embedded in a
> certificate.  Should it say 2 (precertificate) instead?
>
Proposed  fix in
https://github.com/google/certificate-transparency-rfcs/pull/271.

>
> Page 32 refers to the '"submission" output parameter'.  There is no
> submission output parameter.  It should say "submitted_entry" instead.
>
> I think section 1.3 (Major Differences from CT 1.0) should mention the
> replacement of the add-chain and add-pre-chain API endpoints with
> submit-entry.
>
Proposed fix for both in
https://github.com/google/certificate-transparency-rfcs/pull/272.

>
> I am disappointed that the get-sths endpoint isn't included, though I
> accept that this can go in a future document.
>
> A more serious issue is that the signature mutation attack I reported
> is not fixed.  This is a correctness issue and needs to be fixed.  I
> have put a proposed fix out for discussion[1].
>
I'll follow up on that thread.

>
> Regards,
> Andrew
>
> [1] https://www.ietf.org/mail-archive/web/trans/current/msg03035.html
>
> _______________________________________________
> Trans mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/trans
>
_______________________________________________
Trans mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/trans

Reply via email to